Optical critical dimension measurement method, device, medium and equipment

Abstract

The invention provides an optical critical dimension measurement method and device, a medium and equipment. The method comprises the following steps: acquiring a structure parameter set of a grating model and a corresponding first theoretical spectral value; sampling the structural parameter set, and determining a training sample of the spectral value calculation model; and training the spectral value calculation model by taking the output of the spectral value calculation model as a training spectral value and taking the minimum difference between the training spectral value and the corresponding first theoretical spectral value as an optimization target. The trained spectral value calculation model is used as the native database, so that the efficiency of constructing the native database is improved, the size of the native database is reduced, the migration difficulty of the native database is reduced, all structure parameter sets and corresponding spectral values do not need to be loaded when optical critical dimension measurement is carried out, and the measurement efficiency is improved. And the efficiency of optical critical dimension measurement is improved. In the training process of the spectral value calculation model, the acquired grating structure is sampled, so that the time of model training is shortened.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing, and more particularly to an optical critical dimension measurement method, device, medium and equipment. Background Art

[0002] At present, when performing optical critical dimension (OCD) measurement on a grating structure to be measured, the measured spectrum of the grating structure to be measured is usually matched with each theoretical spectrum stored in a pre-built native database, and the structural data corresponding to the matched theoretical spectrum is used as the measured structural data of the grating structure to be measured. Specifically, before performing OCD on the grating to be measured, a grating model corresponding to the grating structure to be measured is usually constructed first, and then the values ​​of various structural parameters of the grating model are determined, and a number of structural points are constructed according to the values ​​of various structural parameters, wherein each structural point contains all kinds of structural parameters, and then the theoretical spectrum corresponding to each structural point is calculated according to the values ​​of the structural parameters contained in each structural point, so as to collect the theoretical spectrum corresponding to each structural point and construct the native database of the grating structure to be measured. In the OCD process, the measured spectrum of the grating structure to be measured is matched with the theoretical spectrum stored in the native database, and the measurement result of the semiconductor to be measured is determined according to the grating structure corresponding to the matched theoretical spectrum. Therefore, the construction of the native database determines the accuracy of the OCD result.

[0003] In the prior art, when constructing a native database, the spectral values ​​corresponding to different structural points are usually calculated by numerical solution methods such as the Finite Element Method (FEM) and the Boundary Element Method (BEM), and then the corresponding spectra are determined based on the determined spectral values. Finally, the native database can be constructed based on the determined spectra and the structural points corresponding to each spectrum. However, as the grating structure of semiconductor chips becomes more and more complex, the amount of spectral value calculation becomes larger and larger, and the calculation time becomes longer and longer, which makes it difficult to meet the needs of users. Therefore, this specification provides an optical critical dimension measurement method, device, medium and equipment. Summary of the invention

[0004] This specification provides an optical key dimension measurement method, device, storage medium and electronic device to partially solve the above-mentioned problems existing in the prior art.

[0005] This manual adopts the following technical solutions: This specification provides an optical critical dimension measurement method, comprising: Acquire a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured; According to the floating value range of the structural parameter and the structural parameter set, sampling is performed to obtain a structural point sampling set; Using the structure point sampling set as training samples, determining the output of the spectrum value calculation model as the training spectrum value; The optimization goal is to minimize the difference between the training spectral value corresponding to the structural point sampling set and the first theoretical spectral value corresponding to the structural point sampling set, and train the spectral value calculation model; wherein the trained spectral value calculation model is used to determine the corresponding second theoretical spectral value according to the input structural parameters.

[0006] Optionally, the step of obtaining a set of structural parameters of the grating model and a floating value range of the structural parameters specifically includes: Setting various structural parameters corresponding to the grating structure according to the grating structure of the grating to be measured; Determining the floating value ranges of various structural parameters according to the manufacturing process of the grating structure; According to the floating value ranges of the various structural parameters, a set of structural parameters of the grating model corresponding to the grating structure is determined.

[0007] Optionally, the step of sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set specifically includes: By means of a random scattering method, at least one value is randomly collected within the floating value range of various structural parameters of the grating model; Constructing each structural point according to the values ​​sampled within the floating value range of the various structural parameters; A structural point sampling set of the grating model is determined according to the structural points.

[0008] Optionally, the step of sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set specifically includes: For each structural parameter, a first sampling step length is determined according to a floating value range of the structural parameter; Sampling within the floating value range according to the first sampling step length to determine each value of the structural parameter; According to the values ​​of various structural parameters obtained by sampling, structural points are constructed, and according to the constructed structural points, a structural point sampling set is determined.

[0009] Optionally, the step of randomly acquiring at least one value within the floating value range of various structural parameters of the grating model specifically includes: For each structural parameter, according to the floating value range of the structural parameter, the floating value range is divided into at least two target ranges; Sampling is performed within each target range to determine at least one value of the structural parameter.

[0010] Optionally, the step of sampling within each target range to determine at least one value of the structural parameter specifically includes: For each target range, determining the number of values ​​of the structural parameter within the target range according to the structural parameter set; Determining a first sampling quantity of the structural parameter according to the quantity of the values, wherein the first sampling quantity is positively correlated with the quantity of the values; The values ​​of the structural parameter of the first sampling quantity are determined within the target range.

[0011] Optionally, the step of sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set specifically includes: The sampling density of the acquisition structure points is adjusted according to the first theoretical spectrum value change rate corresponding to the floating value range of various structural parameters, wherein the sampling density is positively correlated with the first theoretical spectrum value change rate.

[0012] Optionally, the structural points include at least two structural parameters of the side wall angle, height, depth, top width and bottom width of the structure of the grating to be measured, and the spectral value is determined based on the at least two structural parameters and incident light of a preset wavelength.

[0013] Optionally, the spectrum value calculation model is trained through multiple rounds of iterative processes, wherein: The number of structural points in the training samples of each round of iteration is determined in an incremental manner; The structure parameter set is sampled according to the number of structure points in this round of iteration process, and each structure point obtained by sampling is used as a training sample for this round of iteration process.

[0014] Optionally, the spectrum value calculation model is trained through multiple rounds of iterative processes, wherein: Sampling the training samples, and determining at least two training sets containing different numbers of structural points according to the sampled structural points; For each round of iteration, a training set different from the training set used in the previous round of iteration is determined for use in the training of the current round of iteration.

[0015] Optionally, the step of training the spectral value calculation model specifically includes: According to the size of the difference, adjusting the optimization speed / convergence speed of the difference according to a preset training step size; When the number of adjustments reaches a preset value, the preset training step length is reduced, and the model parameters of the spectral value calculation model are adjusted again according to the reduced preset training step length.

[0016] Optionally, the step of training the spectral value calculation model specifically includes: During the training process, recording the differences determined in each round of training of the spectral value calculation model and the model parameters of the spectral value calculation model before the parameters are adjusted in the round of training; When the condition for stopping training is reached, the target difference is determined based on each recorded difference; The model parameters corresponding to the target difference are used as the model parameters of the spectral value calculation model.

[0017] Optionally, the method further comprises: Establishing a theoretical spectrum value database using a plurality of said second theoretical spectrum values; The spectral values ​​of the model are matched with the spectral values ​​in the theoretical spectral database; The structural parameters corresponding to the successfully matched spectral values ​​are used as the structural parameters of the grating to be measured.

[0018] This specification provides an optical critical dimension measuring device, comprising: A construction module, used to obtain a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured; A sampling module, used for sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set; A determination module, used to use the structure point sampling set as a training sample and determine the output of the spectrum value calculation model as a training spectrum value; A training module is used to minimize the difference between the training spectral values ​​corresponding to the structural point sampling set and the first theoretical spectral values ​​corresponding to the structural point sampling set as an optimization target, so as to train the spectral value calculation model; wherein the trained spectral value calculation model is used to calculate the corresponding second theoretical spectral value according to the input structural data.

[0019] The present specification provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, a method for measuring an optical critical dimension is implemented.

[0020] The present specification provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a method for measuring an optical critical dimension is implemented.

[0021] At least one of the above technical solutions adopted in this specification can achieve the following beneficial effects: In a method for measuring an optical critical dimension provided in the present specification, a set of structural parameters of a grating model and a corresponding first theoretical spectrum value are obtained, and the set of structural parameters is sampled to determine a set of structural point samples as training samples of a spectrum value calculation model. The output of the spectrum value calculation model is used as the minimum difference between the training spectrum value and the corresponding first theoretical spectrum value as an optimization goal to train the spectrum value calculation model. The trained spectrum value calculation model is used to determine the corresponding second theoretical spectrum value according to the input structural parameters.

[0022] It can be seen from the above method that the efficiency of building a native database is improved by using a trained spectral value calculation model instead of calculating spectral values ​​through a numerical solution method. In addition, the size of the trained spectral value calculation model is much smaller than the size of the native database, so that when optical critical dimensions are measured on different devices, the trained spectral value calculation model can be directly migrated in different devices, replacing the migration of the native database, which improves the migration speed and saves the storage resources of the device. In addition, during the training process of the spectral value calculation model, the model training time is reduced by sampling the acquired grating structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The illustrative embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification. In the drawings: Figure 1 A schematic diagram of a grating structure provided in this specification; Figure 2 A schematic diagram of a grating structure provided in this specification; Figure 3 A schematic diagram of a spectrum provided in this specification; Figure 4 A schematic diagram of a process of an optical critical dimension measurement method provided in this specification; Figure 5 A schematic diagram of a first theoretical spectrum value change curve provided in this specification; Figure 6 A schematic diagram of a training process of a spectral value calculation model provided in this specification; Figure 7A schematic diagram of an optical critical dimension measurement device provided in this specification; Figure 8 The corresponding Figure 4 Schematic diagram of electronic equipment. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] With the development of semiconductor technology, the manufacturing process of integrated circuits has become more and more sophisticated, the characteristic size of semiconductor chips has become smaller and smaller, and the precision requirements for semiconductor chips have become higher and higher. Small changes in characteristic size will affect the performance and reliability of semiconductor chips. Therefore, it is necessary to accurately measure the characteristic size of semiconductor devices. However, traditional contact measurement methods can no longer meet the measurement accuracy required now, and contact measurement methods may also cause physical damage or contamination to the surface of semiconductor chip samples, resulting in degradation of semiconductor chip performance or even failure. Therefore, non-contact measurement is currently often used to measure the characteristic size of semiconductor chips, namely optical critical dimension measurement (OCD).

[0026] When performing OCD, the optical critical dimensions of the grating structure on the semiconductor chip are usually measured. Figure 1 As shown, Figure 1 A schematic diagram of a grating structure provided in this specification. The grating structure is composed of a plurality of periodically arranged protruding structures. In one or more embodiments of this specification, the grating structure of the grating refers to the cross-sectional shape of the periodically arranged protruding structures constituting the grating, that is, Figure 1 The cross section of each protrusion structure is a trapezoid. Of course, in addition to this, the cross section of the protrusion structure can also be composed of multiple trapezoids, such as Figure 2 As shown, Figure 2 A schematic diagram of a grating structure provided in this specification. Figure 1Some parameters used to describe the characteristics of the cross-section of the raised structure in the grating are marked, such as top width, bottom width, side wall angle, height, depth, etc. These are all structural parameters used to describe the raised structures that constitute the periodic arrangement of the grating. Each raised structure requires at least two structural parameters to be determined. These structural parameters are the optical critical dimensions to be measured in the OCD process. Of course, the specific types of structural parameters to be measured and the number of types of structural parameters to be measured depend on the types of structural parameters contained in the structural points stored in the native database used, and can be selected according to actual needs.

[0027] When performing OCD, currently, the grating to be measured is usually illuminated by light of a preset wavelength at a preset angle to measure the spectrum of the grating to be measured. The measured spectrum is then matched with the spectrum stored in the native database, and the spectrum with the highest similarity to the measured spectrum is determined from the spectrum stored in the native database, and the structural parameters corresponding to the spectrum with the highest similarity are used as the structural parameters of the grating to be measured.

[0028] Therefore, the quality of native database construction determines the accuracy of OCD. At present, when constructing a native database, a grating model is usually pre-constructed, and then several sets of structural parameter values ​​are determined based on the constructed grating model. Several structural points are constructed through permutations and combinations. Each structural point is composed of at least two structural parameters. For example, a grating structure contains three structural parameters, namely top width, height, and sidewall angle. The top width is 3nm, the height is 4nm, and the sidewall angle is 53°. Then a structural point is [3nm, 4nm, 53°]. Then, according to the determined structural points, the spectral value corresponding to each structural point is calculated, and then the corresponding spectrum is determined according to the determined spectral value, wherein the spectral value is a multidimensional characteristic value used to characterize the corresponding spectral feature. For example, the spectral value calculated according to the structural point [3nm, 4nm, 53°] is [0.1533520.123643 0.176632 … 0.45095 0.557975 0.480298]. The multidimensional value as a whole represents a spectral value. The spectrum corresponding to the structural point can be determined according to the multidimensional value, such as Figure 3 As shown, Figure 3 A schematic diagram of a spectrum provided in this specification. Finally, a native database is constructed based on each structural point and the corresponding spectrum.

[0029] Currently, commonly used methods for calculating spectral values ​​include the finite element method and the boundary element method. However, as the grating structure of semiconductor chips becomes more and more complex, the types of structural parameters contained in each structural point are also increasing, resulting in an increasing amount of calculations when calculating spectral values, and the calculation time is also increasing. In order to meet the increasing accuracy requirements of OCD, when building a native database, the change step size between the same type of structural parameters between each structural point is getting smaller and smaller, and the number of structural points is increasing, resulting in a longer and longer time to build the native database, and the memory resources occupied by the native database are also increasing.

[0030] In addition, when performing OCD through different devices, it is necessary to migrate the native database between different devices. However, the native database is getting larger and larger, and the migration time is getting longer and longer, which is difficult to meet the needs of users. Based on this, this specification provides an optical critical dimension measurement method, that is, using a trained spectral value calculation model instead of building a native database to reduce the difficulty of migration and the time consumption of building the native database.

[0031] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0032] like Figure 4 As shown, Figure 4 A schematic diagram of the process of an optical critical dimension measurement method provided in this specification includes the following steps: S300: Acquire a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured.

[0033] In one or more embodiments of this specification, it is not limited to which device specifically performs the optical critical dimension measurement method, such as a mobile terminal and a server. However, since the subsequent steps involve model training, these steps are usually performed by a server. Therefore, this specification will be described by taking the server performing the optical critical dimension measurement method as an example. Among them, the server can be a single device, or composed of multiple devices, such as a distributed server, and this specification does not limit this.

[0034] Since the OCD process relies on optical modeling to match the measurement data, the server can predetermine the structure of the grating to be measured, and then build a grating model based on the structure of the grating to be measured. Of course, in one or more embodiments of this specification, it is not limited to the specific method used by the server to determine the structure of the grating to be measured. It can be determined by the manufacturing process information of the grating to be measured, or it can be determined based on the design file of the grating to be measured, or it can be directly measured by a measuring tool such as a scanning electron microscope to determine the structure of the grating to be measured. Since there are many methods that can be used, this specification will not go into details, and it can be set according to actual needs.

[0035] Then, in order to build a native database more quickly, the server can replace the existing spectral value calculation method by training a spectral value calculation model for spectral value calculation to improve the speed of spectral value calculation. Therefore, before training the spectral value calculation model, the server can first determine the training samples and annotations used to train the spectral value calculation model. When determining the training samples, the server can first obtain the floating value range of the structural parameters of the grating to be measured, the structural parameter set of the grating model, and determine the first theoretical spectral value corresponding to each structural point in the structural parameter set. Among them, the structural parameter set is composed of a number of structural points, each structural point contains at least two structural parameters, and the types of structural parameters contained in each structural point are the same. The floating value range of the structural parameter represents the critical value of the structural parameter.

[0036] Of course, in one or more embodiments of the present specification, it is not limited how the server specifically obtains the floating value range of the structural parameter. When determining the floating value range of the structural parameter, the server can determine the floating value range of the structural parameter according to actual needs. For example, according to expert experience, the theoretical maximum values ​​corresponding to each structural parameter corresponding to the structural parameter are determined, and then the determined theoretical maximum values ​​are used to determine each floating value range, and then a floating value range is randomly selected from each floating value range for the training of the spectrum value calculation model. It is also possible to determine the critical value of the structural parameter according to the critical value of historical data, thereby determining the floating value range of the structural parameter.

[0037] In one or more embodiments of the present specification, the number of determined floating value ranges is not limited, that is, different types of structural parameters can share the same floating value range. For example, the units of the height, depth, top width and bottom width of the structure of the grating to be measured are all nanometers or micrometers. Therefore, [200nm-300nm] can be used as the common floating value range of the height, depth, top width and bottom width of the structure of the grating to be measured. Of course, the floating value range of each structural parameter can also be determined for the structural parameter, which can be set according to actual needs. For example, when the structure of the grating to be measured is more complex, the server can determine the floating value range corresponding to each structural parameter. When the structure of the grating to be measured is relatively simple, the server can use a floating value range as a floating value range shared by multiple structural parameters.

[0038] After determining the floating value range of the structural parameter, the server can obtain the structural parameter combination of the grating model, and then determine the first theoretical spectral value corresponding to the structural point in the native database for each structural point in the structural parameter set.

[0039] It should be noted that there are many methods for the server to obtain the structural parameter set of the grating model. In one or more embodiments of this specification, there is no limitation on this. Other methods except the above methods are described in the subsequent contents of this specification and will not be repeated here.

[0040] In addition, the above-mentioned first theoretical spectrum value is a value used to represent the characteristics of the spectrum corresponding to the grating model. In one or more embodiments of the present specification, the server is not limited to a specific method for determining the first theoretical spectrum value corresponding to the grating model. It can be calculated by numerical solution methods such as Finite Element Method (FEM) and Boundary Element Method (BEM), or it can be calculated by rigorous Rigorous Coupled Wave Analysis (RCWA). The server can be set according to actual needs. This specification will be described later using the calculation of the first theoretical spectrum value by RCWA as an example.

[0041] S302: Sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set.

[0042] Since the number of structure points included in the structure parameter set is usually extremely large, using all the structure points in the structure parameter set as training samples will result in a subsequent long training time for the spectral value calculation model and low training efficiency. In order to reduce the calculation time of the spectral value calculation model, the server can sample the structure point sampling set according to the floating value range of the structure parameter and the structure parameter set, and use the structure point sampling set as the training sample, and the first theoretical spectral value corresponding to the structure point in the structure point sampling set is the label corresponding to the training sample.

[0043] Specifically, the server can determine the maximum value of each structural parameter based on the combination of the structural parameters, and then determine the union of the maximum value of each structural parameter and the floating value range of the structural parameter, sample a preset number of structural points from the determined union through a random scattering method, and then construct a structural point sampling set based on the sampled structural points.

[0044] It should be noted that in one or more embodiments of this specification, there is no restriction on the specific sampling method used by the server to construct the structural point sampling set. Since there are many methods that can be adopted, other methods are described in the subsequent content of this specification and will not be repeated here. Of course, there is no restriction on how the server determines the preset number, which can be set according to actual needs.

[0045] S304: Using the structure point sampling set as training samples, and determining the output of the spectrum value calculation model as the training spectrum value.

[0046] After determining the training samples and annotations for training the spectral value calculation model, the spectral value calculation model to be trained can be trained.

[0047] Specifically, the server inputs the constructed structure point sampling set into the spectral value calculation model to be trained, and determines the output of the spectral value calculation model to be trained as the training spectral value.

[0048] It should be noted that in one or more embodiments of the present specification, it is not limited whether the server inputs all the structural points in the structural point sampling set into the spectral value calculation model to be trained at one time, so that the spectral value calculation model to be trained outputs the training spectral values ​​corresponding to all the structural points, or inputs each structural point into the spectral value settlement model to be trained in batches. It can be set according to actual needs.

[0049] In addition, in one or more embodiments of the present specification, there is no restriction on the specific model used to train the spectral value calculation model to be trained. It can be a convolutional model, a neural network model, etc. The present specification does not impose any restriction on this and can be set according to actual needs. When the spectral value calculation model is obtained through the neural network model, there is no restriction on the way in which the neurons in the spectral value calculation model to be trained are connected. It can be fully connected or non-fully connected and can be set according to actual needs.

[0050] S306: The difference between the training spectral value corresponding to the structural point sampling set and the first theoretical spectral value corresponding to the structural point sampling set is minimized as an optimization target, and the spectral value calculation model is trained; wherein the trained spectral value calculation model is used to determine the corresponding second theoretical spectral value according to the input structural parameters.

[0051] After determining the training spectral values ​​corresponding to each structure point in the structure point sampling set, the server can train the spectral value calculation model to be trained, and use the trained spectral value calculation model as a native database, thereby facilitating the subsequent OCD process. Compared with the traditional native database, the trained spectral value calculation model has a small memory usage and is easy to migrate.

[0052] Specifically, the server determines the first theoretical spectral value corresponding to each structure point in the structure point sampling set, and then determines the difference between the training spectral value corresponding to each structure point and the first theoretical spectral value corresponding to each structure point, and trains the spectral value calculation model to be trained with the minimum difference as the optimization goal.

[0053] After determining the trained spectral value calculation model, the second theoretical spectral value corresponding to each structure point output by the trained spectral value calculation model can be determined according to inputting several groups of structure points into the trained spectral value calculation model, so as to construct a native database according to each second theoretical spectral value and corresponding structure point generated by the spectral value calculation model. Among them, each second theoretical spectral value output by the trained spectral value calculation model is the same as the first theoretical spectral value mentioned above, and is composed of multi-dimensional numerical values. Then determine the spectral value of the measured spectrum, match the spectral value of the measured spectrum with each second theoretical spectral value in the native database, and determine the structure point corresponding to the second theoretical spectral value with the smallest difference as the structure point of the spectral structure of the measured spectrum.

[0054] It should be noted that, in one or more embodiments of the present specification, there is no restriction on whether the native database constructed by the server stores the second theoretical spectral values ​​or the spectra corresponding to the second theoretical spectral values. That is, after determining the second theoretical spectral values ​​output by the trained spectral value calculation model, the native database can be directly constructed based on the structural points input into the trained spectral value calculation model and the second theoretical spectral values ​​output by the trained spectral value calculation model. Then, when performing OCD, the server can convert the measured spectrum into the corresponding measured spectral value, match the measured spectral value with the second theoretical spectral values ​​stored in the native database, determine the second theoretical spectral value with the highest similarity to the measured spectral value, and use the structural point corresponding to the second theoretical spectral value with the highest similarity as the structural point of the spectral structure of the measured spectrum.

[0055] Of course, the server can also determine the spectrum corresponding to each second theoretical spectrum value according to each second theoretical spectrum value after determining each second theoretical spectrum value, and then build a native database based on the spectrum corresponding to the second theoretical spectrum value and the corresponding structure point. When performing OCD, the measured spectrum is matched with each spectrum in the native database to determine the spectrum with the highest similarity to the measured spectrum, and the structure point corresponding to the spectrum with the highest similarity is used as the structure point of the spectrum structure of the measured spectrum. This manual does not limit this, and it can be set according to actual needs.

[0056] based on Figure 4 In the optical critical dimension measurement method, a set of structural parameters of a grating model and a corresponding first theoretical spectrum value are obtained, and the set of structural parameters is sampled to determine a set of structural point samples as training samples of a spectrum value calculation model. The output of the spectrum value calculation model is used as the minimum difference between the training spectrum value and the corresponding first theoretical spectrum value as the optimization goal, and the spectrum value calculation model is trained. The trained spectrum value calculation model is used to determine the corresponding second theoretical spectrum value according to the input structural parameters.

[0057] It can be seen from the above method that using the trained spectral value calculation model as the native database improves the efficiency of building the native database, reduces the size of the native database, and reduces the difficulty of migrating the native database, so that when performing optical critical dimension measurement, there is no need to load all the structural parameter sets and the corresponding spectral values, thereby improving the efficiency of optical critical dimension measurement. During the training process of the spectral value calculation model, the model training time is reduced by sampling the acquired grating structure.

[0058] In order to improve the generalization of the trained spectral value calculation model, that is, the trained spectral value calculation model is applicable to the second theoretical spectral value calculation of structural points with various values, the determined training samples should be reasonable and comprehensive, that is, the value range of each structural parameter should cover all possible variation ranges to ensure that all possible structural features can be measured.

[0059] Therefore, in step S300, when determining the structural parameter set of the grating model, the value ranges of the various structural parameters in the structural parameter set must also be reasonable and comprehensive. Therefore, in order to make the value ranges of the various structural parameters in the structural parameter set reasonable and comprehensive, the server can also set the floating value ranges of the various structural parameters corresponding to the grating structure according to the grating structure of the grating to be measured when determining the structural parameter set of the grating model, and then determine the floating value ranges of the various structural parameters based on the grating structure of the grating to be measured, and then determine the structural parameter set of the grating model of the grating to be measured according to the floating value ranges of the various structural parameters. Of course, if the structural parameter set of the grating model is determined based on the floating value range of the structural parameters. Then, when determining the structural point sampling set, the server can only sample the structural parameter set to determine the structural point sampling set. At this time, the sampled structural points are the structural points in the structural parameter set.

[0060] Specifically, the server first determines the floating value ranges of various structural parameters of the grating structure of the grating to be tested, and then adjusts the values ​​of various structural parameters of the grating model one by one according to the floating value ranges of various structural parameters to determine the structural parameter set of the grating model.

[0061] For example, the grating structure of the grating to be measured includes three structural parameters, among which the floating value range of structural parameter A is (a1~a3), the floating value range of structural parameter B is (b1~b4), and the floating value range of structural parameter C is (c1~c3). When determining the set of structural parameters, the server can first fix the values ​​of structural parameter B and structural parameter C, and then adjust the value of structural parameter A in (a1~a3) to determine several structural points, and then adjust structural parameter B and structural parameter C based on the same idea to determine the structural points with coverage and rationality. Of course, the server can also determine the values ​​of each structural parameter according to the floating value range of various structural parameters, and then arrange and combine the various structural parameters of each value to determine each structural point and construct a set of structural parameters. Of course, the server can also use other methods to determine, and this manual does not limit this, and can be set according to actual needs.

[0062] In addition, when determining the floating value range of the structural parameters, the server may first determine the structural parameter set of the grating model, and then determine the maximum values ​​of various structural parameters according to the structural parameter set, and then determine the floating value ranges of various structural parameters according to the maximum values ​​of various structural parameters. Of course, if the floating value range of the structural parameters is determined based on the structural parameter set. When determining the structural point sampling set, the server may only sample each floating value range, and then arrange and combine the sampled various structural parameters to determine the structural point sampling set. At this time, the sampled structural points may be structural points outside the structural parameter set.

[0063] In addition, the server can also determine the maximum values ​​of various structural parameters in actual manufacturing according to the manufacturing process of the grating structure, and thus determine the floating value range of various structural parameters according to the maximum values ​​of various structural parameters in actual manufacturing. The floating value range determined by this method is often closer to the structural points corresponding to each spectrum in actual application, so the spectral values ​​of the structural points within the floating value range determined by the trained spectral value calculation model can be more accurate.

[0064] Furthermore, when training the spectral value calculation model, in order to prevent the spectral value calculation model from overfitting, the values ​​of each structural parameter of each structural point should be uniformly distributed. Therefore, in step S302, according to the floating value range of the structural parameter and the structural parameter set, when sampling to obtain the structural point sampling set, in order to make the values ​​of each structural parameter of the sampled structural points uniformly distributed within the corresponding floating value range, the server can directly sample the values ​​of various structural parameters in the floating value range of various structural parameters by random scattering method in each determined union, and then arrange and combine the values ​​of various structural parameters to determine the sampled structural points and construct the structural point sampling set.

[0065] It should be noted that when the server determines the values ​​of various structural parameters in each determined union by random scattering, each value is not necessarily an integer, and the value is not necessarily in the structural parameter set. In one or more embodiments of this specification, the server is not limited to how many values ​​are sampled for each structural parameter when sampling. Different types of structural parameters can be sampled to obtain different numbers of values. This specification does not limit this and can be set according to actual needs.

[0066] Furthermore, since when the values ​​of various structural parameters are sampled by the random scattering method, there is also a possibility that the sampled values ​​are non-uniformly distributed in the corresponding union, the server can also determine the corresponding first sampling step according to the union corresponding to each structural parameter, and then perform fixed-step sampling in each determined union according to the corresponding first sampling step, so as to determine the corresponding values ​​of various structural parameters, and then determine each structural point through permutations and combinations, thereby constructing a structural point sampling set.

[0067] It should be noted that in one or more embodiments of the present specification, there is no limitation on the specific method used by the server to determine the first sampling step of various structural parameters. The first sampling step may be pre-set for several times, and then the server matches the pre-set first sampling step with the corresponding floating value range, and determines the corresponding first sampling step according to the matching result.

[0068] For example, the server may pre-set 1-100, a total of 100 first sampling steps, and then determine that the floating value range of a certain structural parameter is 200nm~250nm. In this case, the server may directly ignore the 50 first sampling steps 51-100, and select one from 1-50 according to demand as the first sampling step of this structural parameter.

[0069] The server may also determine the number of values ​​that should be sampled for various structural parameters, and then, for each structural parameter, divide the union corresponding to the structural parameter by the number of values ​​that should be sampled for the structural parameter, thereby determining the first sampling step corresponding to the structural parameter. Of course, the server may also determine the first sampling step corresponding to the various structural parameters by other methods, which is not limited in this specification and may be set according to actual needs.

[0070] In addition, the server may also divide each determined union into at least two target ranges, and determine at least one value of various structural parameters by sampling in each target range, thereby ensuring that the values ​​of various structural parameters are sampled at least within the determined target range. The more target ranges are determined, the closer the values ​​of various structural parameters sampled are to uniform distribution within the corresponding floating value ranges.

[0071] It should be noted that, when the sampled structure point is not in the constructed structure parameter set, the server may determine the first theoretical spectrum value corresponding to the sampled structure point according to RCWA.

[0072] Furthermore, for each structural parameter, when the server samples the number of values ​​of the structural parameter in each target range, the same number of values ​​may be sampled in each target range. However, the distribution probability of the values ​​of the various structural parameters of the grating structure to be measured within the floating value range is often not uniformly distributed. If the values ​​of the various structural parameters of the grating structure to be measured are sampled in large numbers within a certain target range during the training of the spectral value calculation model, the trained spectral value calculation model may be overfitted. Therefore, the server may also determine the first sampling number based on the number of values ​​of the various structural parameters of each structural point in the native database that fall within each target range, wherein the first sampling number is positively correlated with the number of values ​​of the various structural parameters of each structural point in the native database.

[0073] In addition, in order to ensure that the values ​​of the first theoretical spectral values ​​corresponding to each determined structural point are evenly distributed in the value range of each first theoretical spectral value corresponding to the structural parameter set, the server can also adjust the sampling density of the collected structural points according to the change rate of the first theoretical spectral values ​​corresponding to the corresponding union of various structural parameters when sampling the structural parameter set.

[0074] Specifically, the server determines the value range and change rate of each first theoretical spectrum value corresponding to the structure parameter set, and then adjusts the sampling density of the acquisition structure points according to the change rate of the first theoretical spectrum value, wherein the greater the change rate, the greater the sampling density. Figure 5 As shown, Figure 5 A schematic diagram of a first theoretical spectrum value change curve provided in this specification. The vertical axis of the coordinate axis represents the value of the first theoretical spectrum value, the horizontal axis represents the number of the first theoretical spectrum value, and the change curve of the first theoretical spectrum value can be divided into four parts. The corresponding slope of each part represents the change rate k of each first theoretical spectrum value. Since K1>k3>k2=k4, the sampling density corresponding to each part of the curve is z1>z3>z2=z4, that is, for the area where S1 is located, 4 points can be sampled, for the area represented by S3, 3 points can be sampled, and for the area where S2 is located and the area where S4 is located, only one point can be sampled.

[0075] In addition, in step S306, when training the spectral value calculation model to be trained, in order to improve the model accuracy of the spectral value calculation model to be trained and enhance the generalization ability of the spectral value calculation model, the server may also train the spectral value calculation model through multiple rounds of iterations, that is, Figure 6 As shown, Figure 6A schematic diagram of the training process of a spectral value calculation model provided in this specification, after the server samples the structure parameter set, determines the number of training samples to be used in this round of iteration, and then samples the structure parameter set according to the number of training samples to be used in this round of iteration to obtain a structure point sampling set. The sampled structure points are then input into the spectral value calculation model, and the spectral value calculation model is trained with the optimization goal of minimizing the difference between the training spectral value corresponding to the structure point sampling set and the first theoretical spectral value corresponding to the structure point sampling set. After the current round of training is completed, it is determined whether the training end condition of the spectral value calculation model is met. If so, the training is terminated and the spectral value calculation model that has been trained is determined. If not, the number of training samples used in the current round of iteration is determined, and the number of training samples used in the next round of iteration is determined. The structure parameter set is re-sampled to determine the training samples used in the next round of training.

[0076] It should be noted that in one or more embodiments of the present specification, the server is not limited to the specific method used to sample the structure parameter set in each round of training. It can be any of the above-mentioned sampling methods, that is, when sampling the structure parameter set to determine the training samples in each round of training, the sampling method used can be the same or different, and can be set according to actual needs.

[0077] When determining the number of training samples to be used in the next round of iteration according to the number of training samples used in the current round of iteration, the server may use the same number of training samples in each round of training. In order to increase the training efficiency of the model, the server may also use different numbers of training samples in each round of training to train the spectral value calculation model to be trained. For example, the server uses an incremental method to determine the number of training samples in each round, so that in the early stage of model training, the spectral value calculation model to be trained is trained with fewer training samples, which speeds up the convergence of the model and avoids overfitting a small amount of data. Then, as the number of training samples increases in each round, the spectral value calculation model is exposed to more diverse data, thereby improving the generalization ability of the model.

[0078] Specifically, the server determines the number of structure points in the training samples of each round of iteration in an incremental manner, samples the structure parameter set by random scattering according to the number of structure points in the iteration, and uses the sampled structure points as the training samples of the iteration. Of course, the number of structure points used in the next round of iteration can be increased after several rounds of iteration. Or when the duration of training the spectral value calculation model reaches a threshold, the number of structure points used can be increased.

[0079] It should be noted that in one or more embodiments of the present specification, there is no limit on the specific method used by the server to sample the structure parameter set. The server can resample the structure parameter set in each round according to the number of structure points required in the round of training, and use the sampling results as training samples for the current round of training. Before each round of training, the server can resample each structure point in the structure parameter set, and the number of samples can be unlimited. Then, the structure points obtained by this sampling can be used together with the training samples used in the previous round of training as training samples for this round of training. The above two methods can also be combined to alternately determine the training samples required for each round of training. There are many methods that can be used, which can be set according to actual needs.

[0080] When the server determines the number of structural points to be used in the first iterative training, the server may use a preset number as the number of structural points to be used in the first iterative training, or it may be calculated by the multidimensional central limit theorem based on the accuracy and confidence level required by the model, or it may be calculated by other methods, and this manual does not impose any restrictions on this.

[0081] In addition, in addition to determining the number of structural points in the training samples of each iteration process in an incremental manner, the server can also train the spectral value calculation model by alternating the number of samples, that is, through a small number of sample rounds, the convergence of the spectral value calculation model can be accelerated to avoid overfitting, and through a large number of sample rounds, the spectral value calculation model can be globally optimized to improve the generalization ability.

[0082] Specifically, the server can determine the number of structural points to be adopted in the next round based on the number of structural points in the training samples adopted in the current round and the number of structural points in the training samples adopted in the previous round. That is, if the number of structural points adopted in the previous round is greater than the number of structural points adopted in the current round, the number of structural points adopted in the next round is also greater than the number of structural points adopted in the current round. If the number of structural points adopted in the previous round is less than the number of structural points adopted in the current round, the number of structural points adopted in the next round is also less than the number of structural points adopted in the current round. The spectral value calculation model is trained alternately with large and small samples. Of course, the number of structural points adopted in the next round of iteration can also be changed after several rounds of iteration. Or when the duration of training the spectral value calculation model reaches a threshold, the number of structural points adopted can be changed.

[0083] Furthermore, the server may also preset a number of sampling numbers, and then, before each round of iterative training, select a corresponding sampling number from the preset sampling numbers according to the requirements as the number of structural points required for the current round of iterative training. Since there are many methods that can be used, this specification does not limit this, and it can be set according to actual needs.

[0084] Furthermore, when determining the number of structural points in the training samples of each round of iteration, the server may perform sampling once before each round of iteration as described above. Of course, it is also possible to perform sampling of the structural parameter set several times before the training begins, and then determine the training samples used in each round of iteration based on the number of structural points included in each structural point sampling set obtained by sampling.

[0085] That is, after sampling the structure parameter set, the server can also determine at least two training sets containing different numbers of structure points based on the structure points obtained by sampling when determining the structure point sampling set. Then, when training the spectral value calculation model, it is ensured that the training set used in each round of training is different from the training set used in the previous round of training, so as to realize alternating sample quantity training, or the training sets used in every few rounds of training are different from the training sets used in the previous rounds.

[0086] For example, the server divides the sampled structural points into two training sets with different numbers of structural points, wherein the training set with a larger number of structural points is used as a large sample, and the training set with a smaller number of structural points is used as a small sample, and then the large sample is used as a training sample for odd-numbered trainings, and the small sample is used as a training sample for even-numbered trainings. Of course, the large sample can also be used as a training sample for even-numbered trainings, and the small sample can be used as a training sample for odd-numbered trainings.

[0087] In addition, in order to further speed up the training efficiency of the spectral value calculation model, the server can also reduce the training step of the parameters as the number of training times increases when adjusting the parameters of the model, thereby adjusting the optimization speed / convergence speed of the difference, that is, reducing the learning rate of the spectral value calculation model to prevent missing the optimal solution due to excessively long training step length.

[0088] Specifically, the server may first adjust the parameters of the spectral value calculation model according to a preset training step size to reduce the difference between the training spectral value and the corresponding first theoretical spectral value. When the model training reaches a preset value, the training time reaches a preset time, or the size of the difference reaches a preset threshold, the preset training step size is reduced, and the model parameters of the spectral value calculation model are adjusted again according to the reduced preset training step size.

[0089] Furthermore, since the changes in the differences of the spectral value calculation model during the training process are often in an oscillating state, in order to determine the optimal solution of the spectral value calculation model during the training process, the server can also record the differences determined in each round of training of the spectral value calculation model during the training process and the model parameters of the spectral value calculation model before adjusting the parameters in this round of training. When the conditions for stopping training are met, the target difference is determined based on the minimum value of the recorded differences, and the model parameters corresponding to the target difference are used as the model parameters of the spectral value calculation model.

[0090] In addition, it should be noted that in one or more embodiments of the present specification, the trained spectral value calculation model is obtained by training only for one type of structural point, that is, the type and number of structural parameters contained in each input structural point are fixed. When calculating the second theoretical spectral value, the type and number of structural parameters of the input model should be the same as the type and number of structural parameters contained in the structural points used when training the spectral value calculation model. If they are different, a retrained spectral value calculation model is required.

[0091] This specification also provides a corresponding optical critical dimension measuring device, such as Figure 7 As shown, Figure 7 This is a schematic diagram of an optical critical dimension measurement device provided in this specification.

[0092] A construction module 400 is used to obtain a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured; A sampling module 401 is used to obtain a structural point sampling set by sampling according to the floating value range of the structural parameter and the structural parameter set; A determination module 402 is used to use the structure point sampling set as a training sample and determine the output of the spectrum value calculation model as a training spectrum value; The training module 403 is used to train the spectral value calculation model with the optimization goal of minimizing the difference between the training spectral value corresponding to the structural point sampling set and the first theoretical spectral value corresponding to the structural point sampling set; wherein the trained spectral value calculation model is used to calculate the corresponding second theoretical spectral value according to the input structural data.

[0093] Optionally, construction module 400 is used to set various structural parameters corresponding to the grating structure according to the grating structure of the grating to be measured; determine the floating value range of various structural parameters according to the manufacturing process of the grating structure; and determine the structural parameter set of the grating model corresponding to the grating structure according to the floating value range of the various structural parameters.

[0094] Optionally, the sampling module 401 is used to randomly collect at least one value within the floating value range of various structural parameters of the grating model by a random scattering method; construct each structural point according to the values ​​sampled within the floating value range of the various structural parameters; and determine the structural point sampling set of the grating model according to the each structural point.

[0095] Optionally, the sampling module 401 is used to determine a first sampling step for each structural parameter according to a floating value range of the structural parameter; perform sampling within the floating value range according to the first sampling step to determine each value of the structural parameter; construct structural points based on the values ​​of the various structural parameters obtained by sampling, and determine a structural point sampling set based on the constructed structural points.

[0096] Optionally, the sampling module 401 is used to divide the floating value range of each structural parameter into at least two target ranges according to the floating value range of the structural parameter; and sample in each target range to determine at least one value of the structural parameter.

[0097] Optionally, the sampling module 401 is used to determine, for each target range, the number of values ​​of the structural parameter within the target range based on the structural parameter set; determine a first sampling number of the structural parameter based on the number of values, wherein the first sampling number is positively correlated with the number of values; and determine each value of the structural parameter of the first sampling number within the target range.

[0098] Optionally, the sampling module 401 is used to adjust the sampling density of the acquisition structure points according to the first theoretical spectrum value change rate corresponding to the floating value range of various structural parameters, wherein the sampling density is positively correlated with the first theoretical spectrum value change rate.

[0099] Optionally, the spectral value calculation model is trained through multiple rounds of iterative processes; the determination module 402 is used to determine the number of structural points in the training samples of each round of iterative process in an incremental manner; according to the number of structural points in this round of iterative process, the structural parameter set is sampled, and each sampled structural point is used as the training sample of this round of iterative process.

[0100] Optionally, the spectral value calculation model is trained through multiple rounds of iterative processes; a determination module 402 is used to sample the training samples, and determine at least two training sets containing different numbers of structural points based on the sampled structural points; for each round of iterative process, a training set different from the training set used in the previous round of iterative process is determined for use in the training of the current round of iterative process.

[0101] Optionally, the training module 403 is used to adjust the optimization speed / convergence speed of the difference according to the size of the difference according to a preset training step size; when the number of adjustments reaches a preset value, the preset training step size is reduced, and the model parameters of the spectral value calculation model are adjusted again according to the reduced preset training step size.

[0102] Optionally, the training module 403 is used to record the differences determined in each round of training of the spectral value calculation model during the training process and the model parameters of the spectral value calculation model before the parameters are adjusted in this round of training; when the condition for stopping training is met, the target difference is determined according to the recorded differences; and the model parameters corresponding to the target difference are used as the model parameters of the spectral value calculation model.

[0103] Optionally, the device also includes a measurement module 404, which is used to establish a theoretical spectral value database through a number of the second theoretical spectral values; match the spectral values ​​of the grating model to be measured with each spectral value in the theoretical spectral database; and use the structural parameters corresponding to the successfully matched spectral values ​​as the structural parameters of the grating model to be measured.

[0104] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0105] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0106] This specification also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 4 An optical critical dimension measurement method is provided.

[0107] This manual also provides Figure 8As shown in FIG8 , at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 4 Of course, in addition to the software implementation, this specification does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0108] In the 1990s, it was very clear whether the improvement of a technology was hardware improvement (for example, improvement of the circuit structure of diodes, transistors, switches, etc.) or software improvement (improvement of the method flow). However, with the development of technology, many improvements of the method flow today can be regarded as direct improvements of the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be implemented with hardware entity modules. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask chip manufacturers to design and make dedicated integrated circuit chips. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0109] The controller may be implemented in any suitable manner, for example, the controller may take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, and the memory controller may also be implemented as part of the control logic of the memory. It is also known to those skilled in the art that, in addition to implementing the controller in a purely computer-readable program code manner, the controller may be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, such a controller may be considered as a hardware component, and the devices for implementing various functions included therein may also be considered as structures within the hardware component. Or even, the devices for implementing various functions may be considered as both software modules for implementing the method and structures within the hardware component.

[0110] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0111] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0112] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0114] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0116] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0117] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0118] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0119] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0120] It should be understood by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0122] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0123] The above description is only an embodiment of this specification and is not intended to limit this specification. For those skilled in the art, this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the scope of the claims of this application.

Claims

1. An optical critical dimension measurement method, characterized in that: include: Acquire a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured; According to the floating value range of the structural parameter and the structural parameter set, sampling is performed to obtain a structural point sampling set; Using the structure point sampling set as training samples, determining the output of the spectrum value calculation model as the training spectrum value; The spectral value calculation model is trained with the optimization goal of minimizing the difference between the training spectral value corresponding to the structural point sampling set and the first theoretical spectral value corresponding to the structural point sampling set; wherein the trained spectral value calculation model is used to determine the corresponding second theoretical spectral value according to the input structural parameters.

2. The method according to claim 1, characterized in that The step of obtaining the floating value range of the structural parameter specifically includes: Setting various structural parameters corresponding to the grating structure according to the grating structure of the grating to be measured; According to the manufacturing process of the grating structure, the floating value ranges of various structural parameters are determined.

3. The method according to claim 2, characterized in that The step of sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set specifically includes: By means of a random scattering method, at least one value is randomly collected within the floating value range of various structural parameters of the grating model; Constructing each structural point according to the values ​​sampled within the floating value range of the various structural parameters; A structural point sampling set of the grating model is determined according to the structural points.

4. The method according to claim 2, characterized in that The step of sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set specifically includes: For each structural parameter, a first sampling step length is determined according to a floating value range of the structural parameter; Sampling within the floating value range according to the first sampling step length to determine each value of the structural parameter; According to the values ​​of various structural parameters obtained by sampling, structural points are constructed, and according to the constructed structural points, a structural point sampling set is determined.

5. The method according to claim 3, characterized in that The step of randomly collecting at least one value within the floating value range of various structural parameters of the grating model specifically includes: For each structural parameter, according to the floating value range of the structural parameter, the floating value range is divided into at least two target ranges; Sampling is performed within each target range to determine at least one value of the structural parameter.

6. The method according to claim 5, characterized in that The step of sampling within each target range and determining at least one value of the structural parameter specifically includes: For each target range, determining the number of values ​​of the structural parameter within the target range according to the structural parameter set; Determining a first sampling quantity of the structural parameter according to the quantity of the values, wherein the first sampling quantity is positively correlated with the quantity of the values; The values ​​of the structural parameter of the first sampling quantity are determined within the target range.

7. The method according to claim 2, characterized in that According to the floating value range of the structural parameter and the structural parameter set, sampling is performed to obtain a structural point sampling set, specifically including: The sampling density of the acquisition structure points is adjusted according to the change rate of the first theoretical spectrum value corresponding to the floating value range of various structural parameters, wherein the sampling density is positively correlated with the change rate of the first theoretical spectrum value.

8. The method according to claim 1, characterized in that The structural points include at least two structural parameters of the side wall angle, height, depth, top width, and bottom width of the structure of the grating to be measured, and the spectral value is determined based on the at least two structural parameters and incident light of a preset wavelength.

9. The method according to claim 1, characterized in that The spectrum value calculation model is trained through multiple rounds of iterative processes, wherein: The number of structural points in the training samples of each round of iteration is determined in an incremental manner; The structure parameter set is sampled according to the number of structure points in this round of iteration process, and each structure point obtained by sampling is used as a training sample for this round of iteration process.

10. The method according to claim 1, characterized in that The spectrum value calculation model is trained through multiple rounds of iterative processes, wherein: Sampling the training samples, and determining at least two training sets containing different numbers of structural points according to the sampled structural points; For each round of iteration, a training set different from the training set used in the previous round of iteration is determined for use in the training of the current round of iteration.

11. The method according to claim 1, characterized in that The step of training the spectral value calculation model specifically includes: According to the size of the difference, adjusting the optimization speed / convergence speed of the difference according to a preset training step size; When the number of adjustments reaches a preset value, the preset training step length is reduced, and the model parameters of the spectral value calculation model are adjusted again according to the reduced preset training step length.

12. The method according to claim 1, characterized in that The step of training the spectral value calculation model specifically includes: During the training process, recording the differences determined in each round of training of the spectral value calculation model and the model parameters of the spectral value calculation model before the parameters are adjusted in the round of training; When the condition for stopping training is reached, the target difference is determined based on each recorded difference; The model parameters corresponding to the target difference are used as the model parameters of the spectral value calculation model.

13. The method according to claim 1, characterized in that The method further comprises: Establishing a theoretical spectrum value database through a plurality of said second theoretical spectrum values; Matching the spectrum value of the grating to be measured with each spectrum value in the theoretical spectrum database; The structural parameters corresponding to the successfully matched spectral values ​​are used as the structural parameters of the grating to be measured.

14. An optical critical dimension measuring device, characterized in that: include: A construction module, used to obtain a set of structural parameters of a grating model, a floating value range of the structural parameters, and a first theoretical spectrum value corresponding to the grating model, wherein the grating model is established based on the structure of the grating to be measured; A sampling module, used for sampling to obtain a structural point sampling set according to the floating value range of the structural parameter and the structural parameter set; A determination module, used to use the structure point sampling set as a training sample and determine the output of the spectrum value calculation model as a training spectrum value; A training module is used to minimize the difference between the training spectral values ​​corresponding to the structural point sampling set and the first theoretical spectral values ​​corresponding to the structural point sampling set as an optimization target, so as to train the spectral value calculation model; wherein the trained spectral value calculation model is used to calculate the corresponding second theoretical spectral value according to the input structural data.

15. A computer-readable storage medium, characterized in that: The storage medium contains a computer program, and when the computer program is executed by a processor, the method described in any one of claims 1 to 13 is implemented.

16. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 13 when executing the computer program.

Images